Bike-sharing system design problems

Abstract

Bike-sharing systems (BSSs) have emerged in many cities worldwide. One key issue regarding the strategic (long-term) design of conventional station-based BSSs is the deployment of bike stations, as it directly impacts operational efficiency and user experience. Bike station location design is essential since successfully and conveniently picking up and returning shared bikes are the basic requirements for system operation, and have become even more significant with the recent emergence of electric bikes that may require charging stations for recharging. Besides conventional station-based BSSs, innovations in technology have enabled users to lock their bikes to an ordinary bike rack without specific stations, which leads to a free-floating BSS type. Free-floating BSSs allow operators to track the position of each bike using satellite receivers, while users can access bike information through their smartphones using relevant physical and information communication technology infrastructures on bikes. Another important issue that needs to be addressed in tactical (medium-term) BSS design is the static bike relocation, which refers to the task of optimizing the redistribution of bikes among stations in a BSS. Developing appropriate relocation strategies is essential to maintain system balance and availability by reallocating bikes from stations with excess supply to those with higher demand. While there have been extensive studies on the design of BSSs, the following research gaps exist. First, there is a lack of studies that propose a methodological framework for designing the locations of bike stations in a BSS with conventional bikes and e-bikes, which raises a mixed BSS design problem. Second, no studies have provided a methodology to design e-bike sharing systems, particularly the locations and capacity of parking and charging facilities. Such systems incorporate the characteristics of free-floating and station-based systems, which raises a hybrid e-bike sharing system (HEBSS) design problem. Third, lacking studies that propose a methodology for designing a BSS jointly considering the strategic station location design and tactic static bike relocation activity, which raises a combined bike-sharing system (CBSS) design problem. This study aims to fill the research gaps above. A bike station location design problem with conventional bikes and e-bikes to determine the optimal locations of bike stations is first proposed, with the consideration of users’ roaming delay costs. Roaming delay costs refer to the additional costs of a bike user, who is required to roam to another station with available bikes and docks, if he/she finds no shared bikes or docks at a station. Second, a new HEBSS design problem is introduced to determine the optimal locations and capacity of parking and charging facilities. Lastly, a bi-level optimization model is proposed for designing CBSSs that simultaneously considers the optimization design of station location and static bike relocation activity. The problem properties are demonstrated to identify factors that affect the optimal system design and performance, and design insights are provided to obtain a better design for the system operator.